Conventional alarm systems are commonly used to detect and deter intrusions such as break-ins or trespasses. Such alarm systems often employ sensors positioned to monitor a location or region, and trigger an alarm when the sensor detects a break-in or trespass into that region. These sensors are often known sensing elements such as photoelectric beam sensors that detect intrusions when their photoelectric input is disrupted, passive infrared (PIR) sensors that detect infrared radiation emitted by would be intruders, or the like.
While such alarm systems are often effective in detecting and deterring intrusions, they suffer from certain drawbacks. One notable drawback is a propensity for “false alarms.” That is, often the only intrusions of interest are those by persons. Wandering animals, birds, or even falling leaves are often of no cause for concern, yet are commonly detected by an alarm's sensors, triggering an alarm when in fact no real cause for concern exists.
Accordingly, it is desirable to develop alarm systems that have a reduced propensity for detecting false alarms. More specifically, it is desirable to develop alarm systems capable of determining whether a detected intrusion is a potentially undesirable intrusion by a person, or a more harmless intrusion by an animal or inanimate object.